This invention relates to a method of leak testing a package, which comprises placing the package in a sheet-like gas-tight envelope, effecting a differential between the pressure on the outside and the pressure on the inside of the envelope so that the package is tightly enveloped by the envelope, sealing the space ("measuring space") between the package and the envelope pressed against it, and measuring the change in pressure in the sealed measuring space for a predetermined measuring time.
A similar method is disclosed in Aarts, U.S. Pat. No. 4,593,554, issued June 10, 1986. According to that method, a vacuum package is tested for possible leakage by placing the package in a gas-tight and flexible envelope, whereafter pressure is exerted on the outside of the envelope. This pressure serves for causing the envelope to bed down accurately against the circumference of the package, so that only a minute space remains between the envelope and the package. The envelope must not seal a leak which may be present in the package, and, according to the patent application, therefore, the pressure on the envelope should be relatively low. In the space or measuring space referred to a pressure is permitted, normally the atmospheric pressure of the ambient air, which is higher than that in the vacuum package. At that pressure, the measuring space is sealed from the surroundings, whereafter possible changes in pressure within the measuring space are measured for a pre-determined period of time. If there is leak in the vacuum package, these pressure changes will be a decrease in pressure as a result of gas flowing from the measuring space into the vacuum package.
The prior method is limited to the testing of vacuum packages, and cannot be used, without more ado, for testing packages within which the pressure is atmospheric. Summary of the Invention
It is an object of the present invention to provide a method of the above kind, by which, in particular, atmospheric packages can be leak tested.
To that effect, the method according to the invention is characterized in that said pressure differential is so high that the envelope reduces the volume in the package and increases the internal pressure in the non-filled portion of the package, and that a lower pressure than the increased internal pressure in the package is used within the measuring space.
The invention also relates to apparatus for leak testing a package by the above method, comprising a sheet-like gas-tight envelope for receiving the package to be tested, means for supplying a gas or liquid to the outside of the envelope for causing the package to be tightly enveloped by the envelope, sealing means for sealing the space ("measuring space") between the package and the envelope pressed against it from the ambient air, and measuring means for measuring the pressure in the sealed measuring space, characterized in that said means are arranged for the supply of such a pressure that the envelope pressed against the package decreases the free volume of the package and increases the internal pressure within the package, and that means are provided for providing a lower pressure within the measuring space than the increased internal pressure within the package.
For practical reasons, it is desirable for said lower pressure within the measuring space to be the atmospheric pressure of the ambient air. For testing atmospheric packages by the prior method there is then the disadvantage that there is no difference between the pressure within the package and that within the measuring space, so that the presence of a leak cannot be observed by measuring pressure variations within the sealed measuring space. According to the present invention, the envelope is pressed against the package with such a high pressure that the package is slightly compressed, unlike the testing of the vacuum package, in which the envelope just comes to lie in close contact with the package. Owing to the compression of the package, its volume is slightly decreased, resulting in an increase in pressure. In particular in the case of packages in which only a relatively small portion is not filled with product, the increase in pressure can be relatively large with minor compression. As a consequence, a superatmospheric pressure is created within the originally atmospheric package. When the package is leaky, air from the package will flow into the measuring space and there increase the pressure, which is measured as an indication that there is a leak. Accordingly, this flow is contrary to that which occurs in the prior method in leak testing a vacuum package, in which in fact, if the package is leaky, the air will flow from the measuring space into the package.
Although, as stated before, the use of an atmospheric reference pressure in the measuring space is preferred, the difference in pressure between the measuring space and the package may, if desired, be increased further by using a subatmospheric pressure within the measuring space. This feature can be used, for example, if the contents of the package are vulnerable, and hence the package can be compressed to an insufficient extent for creating a sufficiently large pressure differential between package and atmospheric measuring space. In this connection it should be considered that, with the measurement being equally reliable, the measuring time may be shorter when there is a greater pressure differential between package and envelope.
The invention can also be used for testing vacuum packages, in which case the pressure within the sealed measuring space should have a sub-atmospheric value lower than that within the vacuum package. Naturally, packages with a super-atmospheric filling can be tested as well.
Some products are packed in a double-walled package: an inner package, or liner, which is filled with product, and an outer package lying in close contact with the inner package. In such cases, it may be possible for the space between the two packages, which is very small as it is, to be used as the measuring space in the present testing method, in which case this inner space should of course be, or be made, accessible from the outside. As the outer package used as the "envelope" within the meaning of the present invention is already in close contact with the inner package or liner, a relatively slight outer pressure on the outer package will often be all that is necessary for compressing the package. As a consequence, the invention is also effectively usable with such double-walled packages packing fragile products, such as consumption chips.
In view of the nature of the invention, it is in particular suitable for testing air-tight bags containing granular or other loose bulk material, such as non-vacuum packed, ground or non-ground coffee beans, peanuts and cocktail biscuits, and the like, packed in a flexible paper or plastic foil.
Possibly against expectation, even with a relatively high pressure of the envelope against the package, any leaks present therein are not sealed by the envelope. If desired, it can be promoted that any leaks remain open by the selection of the material and the composition of the envelope. Partly in this connection reference is made to the examples of possible envelopes mentioned in the above mentioned U.S. patent of Aarts, which are all also suitable in the present invention. In most cases, a thin-walled, elastic and gas-tight rubber foil will be a very satisfactory envelope. Naturally, it is desirable for the form of the envelope to be more or less adapted to the form of the package to be tested therein. The envelope will mostly be bag-shaped. A bag-shaped envelope is preferably double-walled, with the inner and outer walls of the bag meeting at the open end of the bag,, where they seal the space between the two walls. In this space, a rigid frame may be incorporated, which in essence has the same shape as the bag. This construction of the envelope, described in more detail in European patent application no. 86201251.5, is also used in the apparatus shown in the accompanying drawings.
In some packages, for example, those containing ground coffee, a degasing valve may be provided in the wall of the package. This valve permits the passage of gas in one direction only, namely, from the package to the outside. The escape of gas from the package through the valve during the testing of the package should be prevented. For this purpose, the envelope may be provided with a small thickning or bulge at the proper location, which keeps the valve closed. Alternatively, a sealing ring may be secured to the wall of the envelope at that location, which is pressed by the envelope around the valve against the package, and seals the zone around the valve of the measuring space.
The pressure differential between the outside and the inside of the envelope, required for compressing the package to the desired extent can be determined empirically for each individual case. Mostly, the pressure differential should be greater than about 100 mbar. A pressure differential in the order of 300 mbar will be satisfactory in many cases. Usually, the pressure on the inside of the envelope will be atmospheric, in which case a pressure of 1300 mbar is used on the outside of the envelope. Naturally, as required, any higher or lower suitable pressure differential can be used. The outside pressure can be obtained by supplying a liquid, but preferably by supplying air.
In many cases, the internal pressure within the package during the compression will rise to a value which is equal or substantially equal to the pressure differential acting on the envelope. If, as is commonly the case, the measuring space, before being sealed, is in communication with the atmospheric surroundings, this means that the internal pressure within the atmospheric package can increase to the pressure exerted on the outside of the envelope.
As the measuring space is of very small volume, and is only a fraction of the free space within the package, the increase in pressure within the sealed measuring space as a result of a leak in the package will be relatively large. As a result, a possible leak in the package can commonly be determined within a short measuring period of only a few seconds.